A REVIEW OF THE OPLEGNATHOID FISHES OF JAPAN. 



B}^ David Starr Jordan and Henry W. Fowler, 



Of tlie Leland Stanford Junior Univcrsili/. 



In the present paper is given a review of the Oplegnatkklie found in 

 the waters of Japan. 



Family OPLEGNATHID^. 



Bod}^ rather short and deep, moderately compressed, covered with 

 very small ctenoid scales; lateral line continuous; mouth small, the 

 teeth continuous with the substance of the bones forming- a sharp 

 trenchant edge to the jaws; no teeth on palate; lower pharyngeal 

 l)ones separated; gills 4; 6 or 7 branchiostegals; gill-rakers 20 or 

 21; gill-membranes free from the isthmus; pseudobranchia? large. 

 Dorsal tin with about 12 spines, the spinous portion longer than the 

 soft, the soft fins scaly at base; anal similar to soft dorsal, with 3 

 spines; ventrals thoracic, I, 5; post-temporal forked, not fused with 

 the epiotic. 



A single genus with few species inhabiting the Pacific Ocean, one on 

 the coast of Peru, two in Australia, and two in Japan. The nearest 

 relationships of the family are apparently with the Ila^inulidR' and 

 other Perciform fishes, but it shows some traits of affinity with the 

 Chtetodont forms. The famil}^ difters from all allied forms in the 

 united teeth, which resemble those of the Parrot fishes, although 

 without doubt independently developed. It has no relation to any 

 other group having fused teeth. 



The following notes on the skeleton of Oplegnathus are given us by 

 Mr. Edwin Chapin Starks: 



The genus in certain characters resemble the Ilarclddx (Ephippida'), though it 

 is niucli more typically percoiJ. It is certainly not related to the Cluiiodontidx 

 except as through the Ilarchidx. Post-temporal forked, its lower limb short as in 

 Cltniodipler-us, its upper limb rather loosely overlying the eiiiotic and not attached 

 to it by a suture by its anterior edge. 



The skull is very much like that of Chxtodipterus, though the supraoccipital crest 

 is not so high nor does it extend so far forward. The temporal crest extends farther 

 forward. It does not meet the occipital crest, but stops with it at the frontals. 

 There is a well-developed process descending from the basisphenoid which is absent 

 in the Chsetodipterus. Both have a process from the basisphenoid to which the 



Proceedings U. S. National Museum, Vol. XXV— No. 1278. 



75 



